Professional Experience

  • Present 2020

    Senior Lecturer

    Department of Computer science & Engineering, University of Moratuwa,
    Sri Lanka

  • 2021 2020

    Research Fellow

    LIRNEasia,
    Sri Lanka

  • 2020 2014

    Graduate Research/Teaching Fellow

    University of Oregon, Department of Computer and Information Science,
    USA.

  • 2018 2018

    Givens Associate

    Argonne National Laboratory,
    USA.

  • 2020 2011

    Lecturer

    Department of Computer science & Engineering, University of Moratuwa,
    Sri Lanka

  • 2014 2013

    Researcher

    LIRNEasia,
    Sri Lanka

  • 2014 2013

    Visiting Lecturer

    Northshore College of Business and Technology,
    Sri Lanka

Education

  • Ph.D. 2020

    Ph.D. in Computer & Information Science

    University of Oregon, USA

  • MS 2016

    MS in Computer & Information Science

    University of Oregon, USA

  • BSc2011

    B.Sc Engineering (Hons)in Computer Science & Engineering

    University of Moratuwa, Sri Lanka

Featured Research

Abstract Generation with Hybrid Model Supported by Relevance Matrix


D. Kumarasinghe, and N. de Silva

International Conference on Computational Collective Intelligence, Springer, 2024, pp. 211--223,

In the face of the ever-fast-paced development of natural language processing, the need for the ability to condense information into coherent and concise abstracts is becoming irreplaceable. This study introduces a GPT-Neo-based hybrid model that leverages a relevance matrix for improved summarization of research papers and also stands out for its remarkable resource efficiency. Compared to the most up-to-date state-of-the-art solutions, our model provides a competitive level of performance utilizing minimum computational resources. This efficiency expands the scope of application for such technologies in resource-constrained environments which otherwise would not have been feasible, making them more accessible while eliminating environmental and economic costs. Through detailed methodology and performance evaluation, primarily using ROUGE scores, we show the model’s unique place in keeping a good balance between performance and steadiness. The outcome of our research supports a wider view of NLP development directed at both increased efficiency and accessibility in addition to improvement of the algorithms.